Research On Exercise Prescription Recommendation Method With Ontology Reasoning And Similarity Fusion Calculation

With the progress of society,more and more people pay attention to physical fitness.At present,athletes obtain fitness programs mainly through two ways: fitness coaches and online network platforms.About these two ways,the former has problems such as high cost and poor real-time performance;although athletes can obtain fitness programs anytime and anywhere by the latter,the undifferentiated programs it provides cannot support the individualized sports needs of athletes.In view of this,in this thesis,the exercise prescription knowledge set summarized by sports science,ontology reasoning and similarity fusion calculation method is used to design a sports program recommendation system,which takes into account the individual factors such as real-time state,application intensity,stage goal and so on.The specific research content is as follows:First of all,for the description of the individual characteristics of athletes,this thesis analyzes the exercise prescription formulation process and the needs of the athletes,and divides the characteristics of the athletes into six types: basic information,physical condition,exercise ability,exercise goals,exercise conditions,and exercise preferences,then refine each type to build an athlete model.On this basis,the data required for the model is collected by means of electronic form filling,electronic questionnaire surveys,and system-assisted self-test programs.The electronic form filling method mainly collects information that can be accurately determined by the athletes,and the electronic questionnaire survey mainly collects the relative fuzzy information of athletes,and the auxiliary self-test program mainly collects information on physical fitness and athletic ability that are difficult to determine.Finally,the collected data are analyzed and processed to meet the data requirements of athletes in the exercise prescription recommendation process.Secondly,this thesis designs an exercise prescription recommendation method based on ontology reasoning and similarity fusion calculation.On the one hand,this method uses ontology rule reasoning to determine the exercise prescription parameter constraint space,and the forward reasoning rules adopted are transformed from the general experience of formulating exercise prescriptions in the sports field,so as to ensure the scientificity and rationality of exercise prescription recommendations;On the one hand,in order to solve the problem of the specificity and weak fit of the exercise parameters caused by the large constraint space of the core parameters of the exercise prescription,a method for determining exercise prescription parameters based on similarity fusion calculation is designed.Through the similarity fusion calculation of the highquality cases of the exercise prescription case library,the system can find the personalized prescription parameters with better effects and higher parameter fit in the core parameter constraint space of the exercise prescription,so as to improve the exercise effectiveness of the athletes.At the same time,this thesis further guarantees the rationality,safety and effectiveness of the core parameters of the exercise prescription by designing a level verification strategy for exercise prescription.Finally,this thesis first uses the method of system simulation to analyze the effectiveness of the method.The experimental results show that the recommendation method using similarity fusion calculation has better overall effect and stability in in sports prescription recommendation than the traditional method of direct recommendation based on cooperation.On this basis,based on the research background of mass fitness,a personalized exercise prescription recommendation system was designed and implemented.The exercise prescription was verified through the realization of the exerciser information collection,exercise prescription recommendation,exercise prescription guidance,health management and other functional modules in the system.The availability of recommended models.